From: | Peter Eisentraut <e99re41(at)DoCS(dot)UU(dot)SE> |
---|---|
To: | Zeugswetter Andreas SB <ZeugswetterA(at)wien(dot)spardat(dot)at> |
Cc: | "'hackers(at)postgresql(dot)org'" <hackers(at)postgresql(dot)org> |
Subject: | Re: [HACKERS] Create Group |
Date: | 1999-12-14 11:12:33 |
Message-ID: | Pine.GSO.4.02A.9912141209160.24573-100000@Pingvin.DoCS.UU.SE |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, 14 Dec 1999, Zeugswetter Andreas SB wrote:
> > CREATE GROUP name [ WITH [ SYSID id ] [ USER name1, name2, ... ] ]
> > ALTER GROUP name WITH SYSID id /* changes sysid */
> > ALTER GROUP name ADD USER name1, name2, ...
> > ALTER GROUP name DROP USER name1, name2, ...
> > DROP GROUP name
> I think a group can be interpreted somehow like a priviledge.
> As such the statement to add or remove a user from a group
> would be a "grant" statement.
Not really, at least not in our context. A group is a collection
("group") of users which can collectively be granted privileges. For
example, you can do grant select on your_table to group staff (even right
now).
> The standard mutters something about "role"s
> (again haven't looked it up)
> I don't like the word role instead of group, but maybe if there
> is a standard we should use it.
>
> Informix and Oracle use the keyword role for groups,
> and use grant/revoke to administer them.
I suppose they have a slightly different underlying philosposhy then.
PostgreSQL already uses "group" all over the place, this is just a logical
extension which was missing.
--
Peter Eisentraut Sernanders vaeg 10:115
peter_e(at)gmx(dot)net 75262 Uppsala
http://yi.org/peter-e/ Sweden
From | Date | Subject | |
---|---|---|---|
Next Message | Zeugswetter Andreas SB | 1999-12-14 11:16:24 | AW: [HACKERS] Create Group |
Previous Message | Hiroshi Inoue | 1999-12-14 10:50:23 | Re: AW: [HACKERS] Volunteer: Large Tuples / Tuple chaining |